[PATCH 0/7] seq_printf cleanups

[PATCH 0/7] seq_printf cleanups

[Joe Perches]

seq_printf should return void.

Add a public bool seq_is_full function that can be used to shortcut
unnecesary seq_printf/seq_puts calls when the seq buffer is full.

Start removing the misuses of the seq_printf/seq_puts return value.

Patchset brought forward from an unreplied to set of changes from
back in December 2013.

https://lkml.org/lkml/2013/12/11/713

Renamed seq_is_buf_full to seq_is_full.

Joe Perches (7):
  seq_file: Rename static bool seq_overflow to public bool seq_is_full
  netfilter: Convert print_tuple functions to return void
  dlm: Use seq_is_full - remove seq_printf returns
  dlm: Use seq_puts, remove unnecessary trailing spaces
  fs: Convert show_fdinfo functions to void
  debugfs: Fix misuse of seq_printf return value
  docg3: Fix mixuse of seq_printf return value

 Documentation/filesystems/seq_file.txt       |  28 +--
 Documentation/filesystems/vfs.txt            |   2 +-
 drivers/mtd/devices/docg3.c                  | 112 ++++++------
 drivers/net/tun.c                            |   4 +-
 fs/debugfs/file.c                            |  14 +-
 fs/dlm/debug_fs.c                            | 260 +++++++++++++--------------
 fs/eventfd.c                                 |  15 +-
 fs/eventpoll.c                               |  19 +-
 fs/notify/fdinfo.c                           |  76 ++++----
 fs/notify/fdinfo.h                           |   4 +-
 fs/proc/fd.c                                 |   2 +-
 fs/seq_file.c                                |  28 +--
 fs/signalfd.c                                |  10 +-
 fs/timerfd.c                                 |  27 +--
 include/linux/fs.h                           |   2 +-
 include/linux/seq_file.h                     |   8 +
 include/net/netfilter/nf_conntrack_core.h    |   2 +-
 include/net/netfilter/nf_conntrack_l3proto.h |   4 +-
 include/net/netfilter/nf_conntrack_l4proto.h |   4 +-
 net/netfilter/nf_conntrack_l3proto_generic.c |   5 +-
 net/netfilter/nf_conntrack_proto_dccp.c      |  10 +-
 net/netfilter/nf_conntrack_proto_generic.c   |   5 +-
 net/netfilter/nf_conntrack_proto_gre.c       |  10 +-
 net/netfilter/nf_conntrack_proto_sctp.c      |  10 +-
 net/netfilter/nf_conntrack_proto_tcp.c       |  10 +-
 net/netfilter/nf_conntrack_proto_udp.c       |  10 +-
 net/netfilter/nf_conntrack_proto_udplite.c   |  10 +-
 net/netfilter/nf_conntrack_standalone.c      |  15 +-
 28 files changed, 333 insertions(+), 373 deletions(-)

-- 
1.8.1.2.459.gbcd45b4.dirty

[PATCH 1/7] seq_file: Rename static bool seq_overflow to public bool seq_is_full

[Joe Perches]

The return values of seq_printf/puts/putc are frequently misused.

Start down a path to remove all the return value uses of these
functions.

Make the static bool seq_overflow public along with a rename of
the function to seq_is_full.  Rename the still static
seq_set_overflow to seq_set_full.

Update the documentation to not show return types for seq_printf
et al.  Add a description of seq_is_full.

Signed-off-by: Joe Perches <joe@perches.com>
---
 Documentation/filesystems/seq_file.txt | 28 ++++++++++++++++------------
 fs/seq_file.c                          | 28 ++++++++++++++--------------
 include/linux/seq_file.h               |  8 ++++++++
 3 files changed, 38 insertions(+), 26 deletions(-)

diff --git a/Documentation/filesystems/seq_file.txt b/Documentation/filesystems/seq_file.txt
index 8ea3e90..e19c636 100644
--- a/Documentation/filesystems/seq_file.txt
+++ b/Documentation/filesystems/seq_file.txt
@@ -180,27 +180,23 @@ output must be passed to the seq_file code. Some utility functions have
 been defined which make this task easy.
 
 Most code will simply use seq_printf(), which works pretty much like
-printk(), but which requires the seq_file pointer as an argument. It is
-common to ignore the return value from seq_printf(), but a function
-producing complicated output may want to check that value and quit if
-something non-zero is returned; an error return means that the seq_file
-buffer has been filled and further output will be discarded.
+printk(), but which requires the seq_file pointer as an argument.
 
 For straight character output, the following functions may be used:
 
-	int seq_putc(struct seq_file *m, char c);
-	int seq_puts(struct seq_file *m, const char *s);
-	int seq_escape(struct seq_file *m, const char *s, const char *esc);
+	seq_putc(struct seq_file *m, char c);
+	seq_puts(struct seq_file *m, const char *s);
+	seq_escape(struct seq_file *m, const char *s, const char *esc);
 
 The first two output a single character and a string, just like one would
 expect. seq_escape() is like seq_puts(), except that any character in s
 which is in the string esc will be represented in octal form in the output.
 
-There is also a pair of functions for printing filenames:
+There are also a pair of functions for printing filenames:
 
-	int seq_path(struct seq_file *m, struct path *path, char *esc);
-	int seq_path_root(struct seq_file *m, struct path *path,
-			  struct path *root, char *esc)
+	seq_path(struct seq_file *m, struct path *path, char *esc);
+	seq_path_root(struct seq_file *m, struct path *path,
+		      struct path *root, char *esc)
 
 Here, path indicates the file of interest, and esc is a set of characters
 which should be escaped in the output.  A call to seq_path() will output
@@ -209,6 +205,14 @@ root is desired, it can be used with seq_path_root().  Note that, if it
 turns out that path cannot be reached from root, the value of root will be
 changed in seq_file_root() to a root which *does* work.
 
+A function producing complicated output may want to check
+	bool seq_is_full(struct seq_file *m);
+and avoid further seq_<output> calls if true is returned.
+
+A true return from seq_is_full means that the seq_file buffer is full
+and further output will be discarded.  The seq_show function will attempt
+to allocate a larger buffer and retry printing.
+
 
 Making it all work
 
diff --git a/fs/seq_file.c b/fs/seq_file.c
index 3857b72..555aed6 100644
--- a/fs/seq_file.c
+++ b/fs/seq_file.c
@@ -16,18 +16,18 @@
 #include <asm/uaccess.h>
 #include <asm/page.h>
 
-
 /*
- * seq_files have a buffer which can may overflow. When this happens a larger
+ * seq_files have a buffer which may overflow. When this happens a larger
  * buffer is reallocated and all the data will be printed again.
  * The overflow state is true when m->count == m->size.
  */
-static bool seq_overflow(struct seq_file *m)
+bool seq_is_full(struct seq_file *m)
 {
 	return m->count == m->size;
 }
+EXPORT_SYMBOL(seq_is_full);
 
-static void seq_set_overflow(struct seq_file *m)
+static void seq_set_full(struct seq_file *m)
 {
 	m->count = m->size;
 }
@@ -124,7 +124,7 @@ static int traverse(struct seq_file *m, loff_t offset)
 			error = 0;
 			m->count = 0;
 		}
-		if (seq_overflow(m))
+		if (seq_is_full(m))
 			goto Eoverflow;
 		if (pos + m->count > offset) {
 			m->from = offset - pos;
@@ -267,7 +267,7 @@ Fill:
 			break;
 		}
 		err = m->op->show(m, p);
-		if (seq_overflow(m) || err) {
+		if (seq_is_full(m) || err) {
 			m->count = offs;
 			if (likely(err <= 0))
 				break;
@@ -396,7 +396,7 @@ int seq_escape(struct seq_file *m, const char *s, const char *esc)
 			*p++ = '0' + (c & 07);
 			continue;
 		}
-		seq_set_overflow(m);
+		seq_set_full(m);
 		return -1;
         }
 	m->count = p - m->buf;
@@ -415,7 +415,7 @@ int seq_vprintf(struct seq_file *m, const char *f, va_list args)
 			return 0;
 		}
 	}
-	seq_set_overflow(m);
+	seq_set_full(m);
 	return -1;
 }
 EXPORT_SYMBOL(seq_vprintf);
@@ -557,7 +557,7 @@ int seq_bitmap(struct seq_file *m, const unsigned long *bits,
 			return 0;
 		}
 	}
-	seq_set_overflow(m);
+	seq_set_full(m);
 	return -1;
 }
 EXPORT_SYMBOL(seq_bitmap);
@@ -573,7 +573,7 @@ int seq_bitmap_list(struct seq_file *m, const unsigned long *bits,
 			return 0;
 		}
 	}
-	seq_set_overflow(m);
+	seq_set_full(m);
 	return -1;
 }
 EXPORT_SYMBOL(seq_bitmap_list);
@@ -702,7 +702,7 @@ int seq_puts(struct seq_file *m, const char *s)
 		m->count += len;
 		return 0;
 	}
-	seq_set_overflow(m);
+	seq_set_full(m);
 	return -1;
 }
 EXPORT_SYMBOL(seq_puts);
@@ -736,7 +736,7 @@ int seq_put_decimal_ull(struct seq_file *m, char delimiter,
 	m->count += len;
 	return 0;
 overflow:
-	seq_set_overflow(m);
+	seq_set_full(m);
 	return -1;
 }
 EXPORT_SYMBOL(seq_put_decimal_ull);
@@ -746,7 +746,7 @@ int seq_put_decimal_ll(struct seq_file *m, char delimiter,
 {
 	if (num < 0) {
 		if (m->count + 3 >= m->size) {
-			seq_set_overflow(m);
+			seq_set_full(m);
 			return -1;
 		}
 		if (delimiter)
@@ -774,7 +774,7 @@ int seq_write(struct seq_file *seq, const void *data, size_t len)
 		seq->count += len;
 		return 0;
 	}
-	seq_set_overflow(seq);
+	seq_set_full(seq);
 	return -1;
 }
 EXPORT_SYMBOL(seq_write);
diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h
index 52e0097..1f36da8 100644
--- a/include/linux/seq_file.h
+++ b/include/linux/seq_file.h
@@ -43,6 +43,14 @@ struct seq_operations {
 #define SEQ_SKIP 1
 
 /**
+ * seq_is_full - check if the buffer associated to seq_file is full
+ * @m: the seq_file handle
+ *
+ * Returns true if the buffer is full
+ */
+bool seq_is_full(struct seq_file *m);
+
+/**
  * seq_get_buf - get buffer to write arbitrary data to
  * @m: the seq_file handle
  * @bufp: the beginning of the buffer is stored here
-- 
1.8.1.2.459.gbcd45b4.dirty

Re: [PATCH 1/7] seq_file: Rename static bool seq_overflow to public bool seq_is_full

[Steven Rostedt]

On Mon, 29 Sep 2014 16:08:21 -0700
Joe Perches <joe@perches.com> wrote:

> The return values of seq_printf/puts/putc are frequently misused.
> 
> Start down a path to remove all the return value uses of these
> functions.
> 
> Make the static bool seq_overflow public along with a rename of
> the function to seq_is_full.  Rename the still static
> seq_set_overflow to seq_set_full.
> 
> Update the documentation to not show return types for seq_printf
> et al.  Add a description of seq_is_full.

Actually, can you make a separate function that's public that is
seq_is_full(), where m->count >= m->size, and leave seq_overflow()
alone.

That's because I'm working on making seq_overflow be
m->count > m->size, and allow m->count == m->size not be flagged as an
overflow. I'm looking at places that hard code writing into the buffer
(outside of seq_file.c) and they too can fill the buffer exactly.

Thus, all that is needed is the seq_is_full() function added, and you
don't need to modify the seq_set_overflow() either. It makes sense for
external functions to know test for seq_is_full() as if
m->count == m->size, they can't write anymore either.

Thanks!

-- Steve


> 
> Signed-off-by: Joe Perches <joe@perches.com>
> ---
>  Documentation/filesystems/seq_file.txt | 28 ++++++++++++++++------------
>  fs/seq_file.c                          | 28 ++++++++++++++--------------
>  include/linux/seq_file.h               |  8 ++++++++
>  3 files changed, 38 insertions(+), 26 deletions(-)
> 
> diff --git a/Documentation/filesystems/seq_file.txt b/Documentation/filesystems/seq_file.txt
> index 8ea3e90..e19c636 100644
> --- a/Documentation/filesystems/seq_file.txt
> +++ b/Documentation/filesystems/seq_file.txt
> @@ -180,27 +180,23 @@ output must be passed to the seq_file code. Some utility functions have
>  been defined which make this task easy.
>  
>  Most code will simply use seq_printf(), which works pretty much like
> -printk(), but which requires the seq_file pointer as an argument. It is
> -common to ignore the return value from seq_printf(), but a function
> -producing complicated output may want to check that value and quit if
> -something non-zero is returned; an error return means that the seq_file
> -buffer has been filled and further output will be discarded.
> +printk(), but which requires the seq_file pointer as an argument.
>  
>  For straight character output, the following functions may be used:
>  
> -	int seq_putc(struct seq_file *m, char c);
> -	int seq_puts(struct seq_file *m, const char *s);
> -	int seq_escape(struct seq_file *m, const char *s, const char *esc);
> +	seq_putc(struct seq_file *m, char c);
> +	seq_puts(struct seq_file *m, const char *s);
> +	seq_escape(struct seq_file *m, const char *s, const char *esc);
>  
>  The first two output a single character and a string, just like one would
>  expect. seq_escape() is like seq_puts(), except that any character in s
>  which is in the string esc will be represented in octal form in the output.
>  
> -There is also a pair of functions for printing filenames:
> +There are also a pair of functions for printing filenames:
>  
> -	int seq_path(struct seq_file *m, struct path *path, char *esc);
> -	int seq_path_root(struct seq_file *m, struct path *path,
> -			  struct path *root, char *esc)
> +	seq_path(struct seq_file *m, struct path *path, char *esc);
> +	seq_path_root(struct seq_file *m, struct path *path,
> +		      struct path *root, char *esc)
>  
>  Here, path indicates the file of interest, and esc is a set of characters
>  which should be escaped in the output.  A call to seq_path() will output
> @@ -209,6 +205,14 @@ root is desired, it can be used with seq_path_root().  Note that, if it
>  turns out that path cannot be reached from root, the value of root will be
>  changed in seq_file_root() to a root which *does* work.
>  
> +A function producing complicated output may want to check
> +	bool seq_is_full(struct seq_file *m);
> +and avoid further seq_<output> calls if true is returned.
> +
> +A true return from seq_is_full means that the seq_file buffer is full
> +and further output will be discarded.  The seq_show function will attempt
> +to allocate a larger buffer and retry printing.
> +
>  
>  Making it all work
>  
> diff --git a/fs/seq_file.c b/fs/seq_file.c
> index 3857b72..555aed6 100644
> --- a/fs/seq_file.c
> +++ b/fs/seq_file.c
> @@ -16,18 +16,18 @@
>  #include <asm/uaccess.h>
>  #include <asm/page.h>
>  
> -
>  /*
> - * seq_files have a buffer which can may overflow. When this happens a larger
> + * seq_files have a buffer which may overflow. When this happens a larger
>   * buffer is reallocated and all the data will be printed again.
>   * The overflow state is true when m->count == m->size.
>   */
> -static bool seq_overflow(struct seq_file *m)
> +bool seq_is_full(struct seq_file *m)
>  {
>  	return m->count == m->size;
>  }
> +EXPORT_SYMBOL(seq_is_full);
>  
> -static void seq_set_overflow(struct seq_file *m)
> +static void seq_set_full(struct seq_file *m)
>  {
>  	m->count = m->size;
>  }
> @@ -124,7 +124,7 @@ static int traverse(struct seq_file *m, loff_t offset)
>  			error = 0;
>  			m->count = 0;
>  		}
> -		if (seq_overflow(m))
> +		if (seq_is_full(m))
>  			goto Eoverflow;
>  		if (pos + m->count > offset) {
>  			m->from = offset - pos;
> @@ -267,7 +267,7 @@ Fill:
>  			break;
>  		}
>  		err = m->op->show(m, p);
> -		if (seq_overflow(m) || err) {
> +		if (seq_is_full(m) || err) {
>  			m->count = offs;
>  			if (likely(err <= 0))
>  				break;
> @@ -396,7 +396,7 @@ int seq_escape(struct seq_file *m, const char *s, const char *esc)
>  			*p++ = '0' + (c & 07);
>  			continue;
>  		}
> -		seq_set_overflow(m);
> +		seq_set_full(m);
>  		return -1;
>          }
>  	m->count = p - m->buf;
> @@ -415,7 +415,7 @@ int seq_vprintf(struct seq_file *m, const char *f, va_list args)
>  			return 0;
>  		}
>  	}
> -	seq_set_overflow(m);
> +	seq_set_full(m);
>  	return -1;
>  }
>  EXPORT_SYMBOL(seq_vprintf);
> @@ -557,7 +557,7 @@ int seq_bitmap(struct seq_file *m, const unsigned long *bits,
>  			return 0;
>  		}
>  	}
> -	seq_set_overflow(m);
> +	seq_set_full(m);
>  	return -1;
>  }
>  EXPORT_SYMBOL(seq_bitmap);
> @@ -573,7 +573,7 @@ int seq_bitmap_list(struct seq_file *m, const unsigned long *bits,
>  			return 0;
>  		}
>  	}
> -	seq_set_overflow(m);
> +	seq_set_full(m);
>  	return -1;
>  }
>  EXPORT_SYMBOL(seq_bitmap_list);
> @@ -702,7 +702,7 @@ int seq_puts(struct seq_file *m, const char *s)
>  		m->count += len;
>  		return 0;
>  	}
> -	seq_set_overflow(m);
> +	seq_set_full(m);
>  	return -1;
>  }
>  EXPORT_SYMBOL(seq_puts);
> @@ -736,7 +736,7 @@ int seq_put_decimal_ull(struct seq_file *m, char delimiter,
>  	m->count += len;
>  	return 0;
>  overflow:
> -	seq_set_overflow(m);
> +	seq_set_full(m);
>  	return -1;
>  }
>  EXPORT_SYMBOL(seq_put_decimal_ull);
> @@ -746,7 +746,7 @@ int seq_put_decimal_ll(struct seq_file *m, char delimiter,
>  {
>  	if (num < 0) {
>  		if (m->count + 3 >= m->size) {
> -			seq_set_overflow(m);
> +			seq_set_full(m);
>  			return -1;
>  		}
>  		if (delimiter)
> @@ -774,7 +774,7 @@ int seq_write(struct seq_file *seq, const void *data, size_t len)
>  		seq->count += len;
>  		return 0;
>  	}
> -	seq_set_overflow(seq);
> +	seq_set_full(seq);
>  	return -1;
>  }
>  EXPORT_SYMBOL(seq_write);
> diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h
> index 52e0097..1f36da8 100644
> --- a/include/linux/seq_file.h
> +++ b/include/linux/seq_file.h
> @@ -43,6 +43,14 @@ struct seq_operations {
>  #define SEQ_SKIP 1
>  
>  /**
> + * seq_is_full - check if the buffer associated to seq_file is full
> + * @m: the seq_file handle
> + *
> + * Returns true if the buffer is full
> + */
> +bool seq_is_full(struct seq_file *m);
> +
> +/**
>   * seq_get_buf - get buffer to write arbitrary data to
>   * @m: the seq_file handle
>   * @bufp: the beginning of the buffer is stored here

Re: [PATCH 1/7] seq_file: Rename static bool seq_overflow to public bool seq_is_full

[Joe Perches]

On Mon, 2014-09-29 at 19:44 -0400, Steven Rostedt wrote:
> On Mon, 29 Sep 2014 16:08:21 -0700 Joe Perches <joe@perches.com> wrote:
> > The return values of seq_printf/puts/putc are frequently misused.
> > 
> > Start down a path to remove all the return value uses of these
> > functions.
[]
> Actually, can you make a separate function that's public that is
> seq_is_full(), where m->count >= m->size, and leave seq_overflow()
> alone.

Change the first patch to suit your taste.

The rest of the series should not need change.

Re: [PATCH 1/7] seq_file: Rename static bool seq_overflow to public bool seq_is_full

[Petr Mladek]

On Mon 29-09-14 16:08:21, Joe Perches wrote:
> The return values of seq_printf/puts/putc are frequently misused.
> 
> Start down a path to remove all the return value uses of these
> functions.
> 
> Make the static bool seq_overflow public along with a rename of
> the function to seq_is_full.  Rename the still static
> seq_set_overflow to seq_set_full.
> 
> Update the documentation to not show return types for seq_printf
> et al.  Add a description of seq_is_full.
> 
> Signed-off-by: Joe Perches <joe@perches.com>
> ---
>  Documentation/filesystems/seq_file.txt | 28 ++++++++++++++++------------
>  fs/seq_file.c                          | 28 ++++++++++++++--------------
>  include/linux/seq_file.h               |  8 ++++++++
>  3 files changed, 38 insertions(+), 26 deletions(-)
> 
> diff --git a/Documentation/filesystems/seq_file.txt b/Documentation/filesystems/seq_file.txt
> index 8ea3e90..e19c636 100644
> --- a/Documentation/filesystems/seq_file.txt
> +++ b/Documentation/filesystems/seq_file.txt
> @@ -180,27 +180,23 @@ output must be passed to the seq_file code. Some utility functions have
>  been defined which make this task easy.
>  
>  Most code will simply use seq_printf(), which works pretty much like
> -printk(), but which requires the seq_file pointer as an argument. It is
> -common to ignore the return value from seq_printf(), but a function
> -producing complicated output may want to check that value and quit if
> -something non-zero is returned; an error return means that the seq_file
> -buffer has been filled and further output will be discarded.
> +printk(), but which requires the seq_file pointer as an argument.
>  
>  For straight character output, the following functions may be used:
>  
> -	int seq_putc(struct seq_file *m, char c);
> -	int seq_puts(struct seq_file *m, const char *s);
> -	int seq_escape(struct seq_file *m, const char *s, const char *esc);
> +	seq_putc(struct seq_file *m, char c);
> +	seq_puts(struct seq_file *m, const char *s);
> +	seq_escape(struct seq_file *m, const char *s, const char *esc);
>  
>  The first two output a single character and a string, just like one would
>  expect. seq_escape() is like seq_puts(), except that any character in s
>  which is in the string esc will be represented in octal form in the output.
>  
> -There is also a pair of functions for printing filenames:
> +There are also a pair of functions for printing filenames:
>  
> -	int seq_path(struct seq_file *m, struct path *path, char *esc);
> -	int seq_path_root(struct seq_file *m, struct path *path,
> -			  struct path *root, char *esc)
> +	seq_path(struct seq_file *m, struct path *path, char *esc);
> +	seq_path_root(struct seq_file *m, struct path *path,
> +		      struct path *root, char *esc)
>  
>  Here, path indicates the file of interest, and esc is a set of characters
>  which should be escaped in the output.  A call to seq_path() will output
> @@ -209,6 +205,14 @@ root is desired, it can be used with seq_path_root().  Note that, if it
>  turns out that path cannot be reached from root, the value of root will be
>  changed in seq_file_root() to a root which *does* work.
>  
> +A function producing complicated output may want to check
> +	bool seq_is_full(struct seq_file *m);
> +and avoid further seq_<output> calls if true is returned.
> +
> +A true return from seq_is_full means that the seq_file buffer is full
> +and further output will be discarded.  The seq_show function will attempt
> +to allocate a larger buffer and retry printing.
> +
>  
>  Making it all work
>  
> diff --git a/fs/seq_file.c b/fs/seq_file.c
> index 3857b72..555aed6 100644
> --- a/fs/seq_file.c
> +++ b/fs/seq_file.c
> @@ -16,18 +16,18 @@
>  #include <asm/uaccess.h>
>  #include <asm/page.h>
>  
> -
>  /*
> - * seq_files have a buffer which can may overflow. When this happens a larger
> + * seq_files have a buffer which may overflow. When this happens a larger
>   * buffer is reallocated and all the data will be printed again.
>   * The overflow state is true when m->count == m->size.
>   */
> -static bool seq_overflow(struct seq_file *m)
> +bool seq_is_full(struct seq_file *m)
>  {
>  	return m->count == m->size;
>  }
> +EXPORT_SYMBOL(seq_is_full);
>  
> -static void seq_set_overflow(struct seq_file *m)
> +static void seq_set_full(struct seq_file *m)
>  {
>  	m->count = m->size;
>  }
> @@ -124,7 +124,7 @@ static int traverse(struct seq_file *m, loff_t offset)
>  			error = 0;
>  			m->count = 0;
>  		}
> -		if (seq_overflow(m))
> +		if (seq_is_full(m))
>  			goto Eoverflow;

I would keep seq_overflow() here. seq_is_full() should mean that the
data still fit into the buffer.

>  		if (pos + m->count > offset) {
>  			m->from = offset - pos;
> @@ -267,7 +267,7 @@ Fill:
>  			break;
>  		}
>  		err = m->op->show(m, p);
> -		if (seq_overflow(m) || err) {
> +		if (seq_is_full(m) || err) {

same here

>  			m->count = offs;
>  			if (likely(err <= 0))
>  				break;
> @@ -396,7 +396,7 @@ int seq_escape(struct seq_file *m, const char *s, const char *esc)
>  			*p++ = '0' + (c & 07);
>  			continue;
>  		}
> -		seq_set_overflow(m);
> +		seq_set_full(m);

I would keep seq_set_overflow() here because the data did not fit in.

>  		return -1;
>          }
>  	m->count = p - m->buf;
> @@ -415,7 +415,7 @@ int seq_vprintf(struct seq_file *m, const char *f, va_list args)
>  			return 0;
>  		}
>  	}
> -	seq_set_overflow(m);
> +	seq_set_full(m);

Hmm, we do not know if the data are shrunken by vsnprintf(). I would
keep seq_set_overflow() here to be on the safe side.

>  	return -1;
>  }
>  EXPORT_SYMBOL(seq_vprintf);
> @@ -557,7 +557,7 @@ int seq_bitmap(struct seq_file *m, const unsigned long *bits,
>  			return 0;
>  		}
>  	}
> -	seq_set_overflow(m);
> +	seq_set_full(m);

same here

>  	return -1;
>  }
>  EXPORT_SYMBOL(seq_bitmap);
> @@ -573,7 +573,7 @@ int seq_bitmap_list(struct seq_file *m, const unsigned long *bits,
>  			return 0;
>  		}
>  	}
> -	seq_set_overflow(m);
> +	seq_set_full(m);

and here

>  	return -1;
>  }
>  EXPORT_SYMBOL(seq_bitmap_list);
> @@ -702,7 +702,7 @@ int seq_puts(struct seq_file *m, const char *s)
>  		m->count += len;
>  		return 0;
>  	}
> -	seq_set_overflow(m);
> +	seq_set_full(m);

and here

>  	return -1;
>  }
>  EXPORT_SYMBOL(seq_puts);
> @@ -736,7 +736,7 @@ int seq_put_decimal_ull(struct seq_file *m, char delimiter,
>  	m->count += len;
>  	return 0;
>  overflow:
> -	seq_set_overflow(m);
> +	seq_set_full(m);

We should change one above contition from

	if (m->count + 2 >= m->size) /* we'll write 2 bytes at least */
		goto overflow;

to

	if (m->count + 2 > m->size) /* we'll write 2 bytes at least */
		goto overflow;

>  	return -1;
>  }
>  EXPORT_SYMBOL(seq_put_decimal_ull);
> @@ -746,7 +746,7 @@ int seq_put_decimal_ll(struct seq_file *m, char delimiter,
>  {
>  	if (num < 0) {
>  		if (m->count + 3 >= m->size) {
> -			seq_set_overflow(m);
> +			seq_set_full(m);
>  			return -1;

We should change this to:

		if (m->count + 3 > m->size) {
			seq_set_overflow(m);


>  		}
>  		if (delimiter)
> @@ -774,7 +774,7 @@ int seq_write(struct seq_file *seq, const void *data, size_t len)
>  		seq->count += len;
>  		return 0;
>  	}
> -	seq_set_overflow(seq);
> +	seq_set_full(seq);

IMHO, the whole function should be:

void seq_write(struct seq_file *seq, const void *data, size_t len)
{
	if (seq->count + len <= seq->size) {
		memcpy(seq->buf + seq->count, data, len);
		seq->count += len;
		return 0;
	}
	seq_set_overflow(seq);
}

I mean that the overflow should be set only when there is a real overflow.


All in all, this patch should depend on the Steven's work and most of
it will be unused.

Best Regards,
Petr

>  	return -1;
>  }
>  EXPORT_SYMBOL(seq_write);
> diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h
> index 52e0097..1f36da8 100644
> --- a/include/linux/seq_file.h
> +++ b/include/linux/seq_file.h
> @@ -43,6 +43,14 @@ struct seq_operations {
>  #define SEQ_SKIP 1
>  
>  /**
> + * seq_is_full - check if the buffer associated to seq_file is full
> + * @m: the seq_file handle
> + *
> + * Returns true if the buffer is full
> + */
> +bool seq_is_full(struct seq_file *m);
> +
> +/**
>   * seq_get_buf - get buffer to write arbitrary data to
>   * @m: the seq_file handle
>   * @bufp: the beginning of the buffer is stored here
> -- 
> 1.8.1.2.459.gbcd45b4.dirty
> 

[PATCH 5/7] fs: Convert show_fdinfo functions to void

[Joe Perches]

seq_printf functions shouldn't really check the return value.
Checking seq_is_full occasionally is used instead.

Update vfs documentation.

Signed-off-by: Joe Perches <joe@perches.com>
---
 Documentation/filesystems/vfs.txt |  2 +-
 drivers/net/tun.c                 |  4 +--
 fs/eventfd.c                      | 15 +++-----
 fs/eventpoll.c                    | 19 ++++------
 fs/notify/fdinfo.c                | 76 +++++++++++++++++----------------------
 fs/notify/fdinfo.h                |  4 +--
 fs/proc/fd.c                      |  2 +-
 fs/signalfd.c                     | 10 ++----
 fs/timerfd.c                      | 27 +++++++-------
 include/linux/fs.h                |  2 +-
 10 files changed, 68 insertions(+), 93 deletions(-)

diff --git a/Documentation/filesystems/vfs.txt b/Documentation/filesystems/vfs.txt
index fceff7c..af1dbc1 100644
--- a/Documentation/filesystems/vfs.txt
+++ b/Documentation/filesystems/vfs.txt
@@ -828,7 +828,7 @@ struct file_operations {
 	ssize_t (*splice_read)(struct file *, struct pipe_inode_info *, size_t, unsigned int);
 	int (*setlease)(struct file *, long arg, struct file_lock **, void **);
 	long (*fallocate)(struct file *, int mode, loff_t offset, loff_t len);
-	int (*show_fdinfo)(struct seq_file *m, struct file *f);
+	void (*show_fdinfo)(struct seq_file *m, struct file *f);
 };
 
 Again, all methods are called without any locks being held, unless
diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index 186ce54..a3420e0 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -2209,7 +2209,7 @@ static int tun_chr_close(struct inode *inode, struct file *file)
 }
 
 #ifdef CONFIG_PROC_FS
-static int tun_chr_show_fdinfo(struct seq_file *m, struct file *f)
+static void tun_chr_show_fdinfo(struct seq_file *m, struct file *f)
 {
 	struct tun_struct *tun;
 	struct ifreq ifr;
@@ -2225,7 +2225,7 @@ static int tun_chr_show_fdinfo(struct seq_file *m, struct file *f)
 	if (tun)
 		tun_put(tun);
 
-	return seq_printf(m, "iff:\t%s\n", ifr.ifr_name);
+	seq_printf(m, "iff:\t%s\n", ifr.ifr_name);
 }
 #endif
 
diff --git a/fs/eventfd.c b/fs/eventfd.c
index d6a88e7..abcb25d 100644
--- a/fs/eventfd.c
+++ b/fs/eventfd.c
@@ -286,25 +286,20 @@ static ssize_t eventfd_write(struct file *file, const char __user *buf, size_t c
 	return res;
 }
 
-#ifdef CONFIG_PROC_FS
-static int eventfd_show_fdinfo(struct seq_file *m, struct file *f)
+static void eventfd_show_fdinfo(struct seq_file *m, struct file *f)
 {
+#ifdef CONFIG_PROC_FS
 	struct eventfd_ctx *ctx = f->private_data;
-	int ret;
 
 	spin_lock_irq(&ctx->wqh.lock);
-	ret = seq_printf(m, "eventfd-count: %16llx\n",
-			 (unsigned long long)ctx->count);
+	seq_printf(m, "eventfd-count: %16llx\n",
+		   (unsigned long long)ctx->count);
 	spin_unlock_irq(&ctx->wqh.lock);
-
-	return ret;
-}
 #endif
+}
 
 static const struct file_operations eventfd_fops = {
-#ifdef CONFIG_PROC_FS
 	.show_fdinfo	= eventfd_show_fdinfo,
-#endif
 	.release	= eventfd_release,
 	.poll		= eventfd_poll,
 	.read		= eventfd_read,
diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index 7bcfff9..4e742b6 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -869,34 +869,29 @@ static unsigned int ep_eventpoll_poll(struct file *file, poll_table *wait)
 	return pollflags != -1 ? pollflags : 0;
 }
 
-#ifdef CONFIG_PROC_FS
-static int ep_show_fdinfo(struct seq_file *m, struct file *f)
+static void ep_show_fdinfo(struct seq_file *m, struct file *f)
 {
+#ifdef CONFIG_PROC_FS
 	struct eventpoll *ep = f->private_data;
 	struct rb_node *rbp;
-	int ret = 0;
 
 	mutex_lock(&ep->mtx);
 	for (rbp = rb_first(&ep->rbr); rbp; rbp = rb_next(rbp)) {
 		struct epitem *epi = rb_entry(rbp, struct epitem, rbn);
 
-		ret = seq_printf(m, "tfd: %8d events: %8x data: %16llx\n",
-				 epi->ffd.fd, epi->event.events,
-				 (long long)epi->event.data);
-		if (ret)
+		seq_printf(m, "tfd: %8d events: %8x data: %16llx\n",
+			   epi->ffd.fd, epi->event.events,
+			   (long long)epi->event.data);
+		if (seq_is_full(m))
 			break;
 	}
 	mutex_unlock(&ep->mtx);
-
-	return ret;
-}
 #endif
+}
 
 /* File callbacks that implement the eventpoll file behaviour */
 static const struct file_operations eventpoll_fops = {
-#ifdef CONFIG_PROC_FS
 	.show_fdinfo	= ep_show_fdinfo,
-#endif
 	.release	= ep_eventpoll_release,
 	.poll		= ep_eventpoll_poll,
 	.llseek		= noop_llseek,
diff --git a/fs/notify/fdinfo.c b/fs/notify/fdinfo.c
index 9d7e2b9..0bb10b7 100644
--- a/fs/notify/fdinfo.c
+++ b/fs/notify/fdinfo.c
@@ -20,25 +20,24 @@
 
 #if defined(CONFIG_INOTIFY_USER) || defined(CONFIG_FANOTIFY)
 
-static int show_fdinfo(struct seq_file *m, struct file *f,
-		       int (*show)(struct seq_file *m, struct fsnotify_mark *mark))
+static void show_fdinfo(struct seq_file *m, struct file *f,
+			void (*show)(struct seq_file *m,
+				     struct fsnotify_mark *mark))
 {
 	struct fsnotify_group *group = f->private_data;
 	struct fsnotify_mark *mark;
-	int ret = 0;
 
 	mutex_lock(&group->mark_mutex);
 	list_for_each_entry(mark, &group->marks_list, g_list) {
-		ret = show(m, mark);
-		if (ret)
+		show(m, mark);
+		if (seq_is_full(m))
 			break;
 	}
 	mutex_unlock(&group->mark_mutex);
-	return ret;
 }
 
 #if defined(CONFIG_EXPORTFS)
-static int show_mark_fhandle(struct seq_file *m, struct inode *inode)
+static void show_mark_fhandle(struct seq_file *m, struct inode *inode)
 {
 	struct {
 		struct file_handle handle;
@@ -52,71 +51,62 @@ static int show_mark_fhandle(struct seq_file *m, struct inode *inode)
 	ret = exportfs_encode_inode_fh(inode, (struct fid *)f.handle.f_handle, &size, 0);
 	if ((ret == FILEID_INVALID) || (ret < 0)) {
 		WARN_ONCE(1, "Can't encode file handler for inotify: %d\n", ret);
-		return 0;
+		return;
 	}
 
 	f.handle.handle_type = ret;
 	f.handle.handle_bytes = size * sizeof(u32);
 
-	ret = seq_printf(m, "fhandle-bytes:%x fhandle-type:%x f_handle:",
-			 f.handle.handle_bytes, f.handle.handle_type);
+	seq_printf(m, "fhandle-bytes:%x fhandle-type:%x f_handle:",
+		   f.handle.handle_bytes, f.handle.handle_type);
 
 	for (i = 0; i < f.handle.handle_bytes; i++)
-		ret |= seq_printf(m, "%02x", (int)f.handle.f_handle[i]);
-
-	return ret;
+		seq_printf(m, "%02x", (int)f.handle.f_handle[i]);
 }
 #else
-static int show_mark_fhandle(struct seq_file *m, struct inode *inode)
+static void show_mark_fhandle(struct seq_file *m, struct inode *inode)
 {
-	return 0;
 }
 #endif
 
 #ifdef CONFIG_INOTIFY_USER
 
-static int inotify_fdinfo(struct seq_file *m, struct fsnotify_mark *mark)
+static void inotify_fdinfo(struct seq_file *m, struct fsnotify_mark *mark)
 {
 	struct inotify_inode_mark *inode_mark;
 	struct inode *inode;
-	int ret = 0;
 
 	if (!(mark->flags & (FSNOTIFY_MARK_FLAG_ALIVE | FSNOTIFY_MARK_FLAG_INODE)))
-		return 0;
+		return;
 
 	inode_mark = container_of(mark, struct inotify_inode_mark, fsn_mark);
 	inode = igrab(mark->i.inode);
 	if (inode) {
-		ret = seq_printf(m, "inotify wd:%x ino:%lx sdev:%x "
-				 "mask:%x ignored_mask:%x ",
-				 inode_mark->wd, inode->i_ino,
-				 inode->i_sb->s_dev,
-				 mark->mask, mark->ignored_mask);
-		ret |= show_mark_fhandle(m, inode);
-		ret |= seq_putc(m, '\n');
+		seq_printf(m, "inotify wd:%x ino:%lx sdev:%x mask:%x ignored_mask:%x ",
+			   inode_mark->wd, inode->i_ino, inode->i_sb->s_dev,
+			   mark->mask, mark->ignored_mask);
+		show_mark_fhandle(m, inode);
+		seq_putc(m, '\n');
 		iput(inode);
 	}
-
-	return ret;
 }
 
-int inotify_show_fdinfo(struct seq_file *m, struct file *f)
+void inotify_show_fdinfo(struct seq_file *m, struct file *f)
 {
-	return show_fdinfo(m, f, inotify_fdinfo);
+	show_fdinfo(m, f, inotify_fdinfo);
 }
 
 #endif /* CONFIG_INOTIFY_USER */
 
 #ifdef CONFIG_FANOTIFY
 
-static int fanotify_fdinfo(struct seq_file *m, struct fsnotify_mark *mark)
+static void fanotify_fdinfo(struct seq_file *m, struct fsnotify_mark *mark)
 {
 	unsigned int mflags = 0;
 	struct inode *inode;
-	int ret = 0;
 
 	if (!(mark->flags & FSNOTIFY_MARK_FLAG_ALIVE))
-		return 0;
+		return;
 
 	if (mark->flags & FSNOTIFY_MARK_FLAG_IGNORED_SURV_MODIFY)
 		mflags |= FAN_MARK_IGNORED_SURV_MODIFY;
@@ -125,25 +115,23 @@ static int fanotify_fdinfo(struct seq_file *m, struct fsnotify_mark *mark)
 		inode = igrab(mark->i.inode);
 		if (!inode)
 			goto out;
-		ret = seq_printf(m, "fanotify ino:%lx sdev:%x "
-				 "mflags:%x mask:%x ignored_mask:%x ",
-				 inode->i_ino, inode->i_sb->s_dev,
-				 mflags, mark->mask, mark->ignored_mask);
-		ret |= show_mark_fhandle(m, inode);
-		ret |= seq_putc(m, '\n');
+		seq_printf(m, "fanotify ino:%lx sdev:%x mflags:%x mask:%x ignored_mask:%x ",
+			   inode->i_ino, inode->i_sb->s_dev,
+			   mflags, mark->mask, mark->ignored_mask);
+		show_mark_fhandle(m, inode);
+		seq_putc(m, '\n');
 		iput(inode);
 	} else if (mark->flags & FSNOTIFY_MARK_FLAG_VFSMOUNT) {
 		struct mount *mnt = real_mount(mark->m.mnt);
 
-		ret = seq_printf(m, "fanotify mnt_id:%x mflags:%x mask:%x "
-				 "ignored_mask:%x\n", mnt->mnt_id, mflags,
-				 mark->mask, mark->ignored_mask);
+		seq_printf(m, "fanotify mnt_id:%x mflags:%x mask:%x ignored_mask:%x\n",
+			   mnt->mnt_id, mflags, mark->mask, mark->ignored_mask);
 	}
 out:
-	return ret;
+	;
 }
 
-int fanotify_show_fdinfo(struct seq_file *m, struct file *f)
+void fanotify_show_fdinfo(struct seq_file *m, struct file *f)
 {
 	struct fsnotify_group *group = f->private_data;
 	unsigned int flags = 0;
@@ -169,7 +157,7 @@ int fanotify_show_fdinfo(struct seq_file *m, struct file *f)
 	seq_printf(m, "fanotify flags:%x event-flags:%x\n",
 		   flags, group->fanotify_data.f_flags);
 
-	return show_fdinfo(m, f, fanotify_fdinfo);
+	show_fdinfo(m, f, fanotify_fdinfo);
 }
 
 #endif /* CONFIG_FANOTIFY */
diff --git a/fs/notify/fdinfo.h b/fs/notify/fdinfo.h
index 556afda..9664c49 100644
--- a/fs/notify/fdinfo.h
+++ b/fs/notify/fdinfo.h
@@ -10,11 +10,11 @@ struct file;
 #ifdef CONFIG_PROC_FS
 
 #ifdef CONFIG_INOTIFY_USER
-extern int inotify_show_fdinfo(struct seq_file *m, struct file *f);
+void inotify_show_fdinfo(struct seq_file *m, struct file *f);
 #endif
 
 #ifdef CONFIG_FANOTIFY
-extern int fanotify_show_fdinfo(struct seq_file *m, struct file *f);
+void fanotify_show_fdinfo(struct seq_file *m, struct file *f);
 #endif
 
 #else /* CONFIG_PROC_FS */
diff --git a/fs/proc/fd.c b/fs/proc/fd.c
index e11d7c5..4c3c253 100644
--- a/fs/proc/fd.c
+++ b/fs/proc/fd.c
@@ -53,7 +53,7 @@ static int seq_show(struct seq_file *m, void *v)
 			   (long long)file->f_pos, f_flags,
 			   real_mount(file->f_path.mnt)->mnt_id);
 		if (file->f_op->show_fdinfo)
-			ret = file->f_op->show_fdinfo(m, file);
+			file->f_op->show_fdinfo(m, file);
 		fput(file);
 	}
 
diff --git a/fs/signalfd.c b/fs/signalfd.c
index 424b7b6..06bd5a2 100644
--- a/fs/signalfd.c
+++ b/fs/signalfd.c
@@ -229,24 +229,20 @@ static ssize_t signalfd_read(struct file *file, char __user *buf, size_t count,
 	return total ? total: ret;
 }
 
-#ifdef CONFIG_PROC_FS
-static int signalfd_show_fdinfo(struct seq_file *m, struct file *f)
+static void signalfd_show_fdinfo(struct seq_file *m, struct file *f)
 {
+#ifdef CONFIG_PROC_FS
 	struct signalfd_ctx *ctx = f->private_data;
 	sigset_t sigmask;
 
 	sigmask = ctx->sigmask;
 	signotset(&sigmask);
 	render_sigset_t(m, "sigmask:\t", &sigmask);
-
-	return 0;
-}
 #endif
+}
 
 static const struct file_operations signalfd_fops = {
-#ifdef CONFIG_PROC_FS
 	.show_fdinfo	= signalfd_show_fdinfo,
-#endif
 	.release	= signalfd_release,
 	.poll		= signalfd_poll,
 	.read		= signalfd_read,
diff --git a/fs/timerfd.c b/fs/timerfd.c
index b46ffa9..b94fa6c 100644
--- a/fs/timerfd.c
+++ b/fs/timerfd.c
@@ -288,7 +288,7 @@ static ssize_t timerfd_read(struct file *file, char __user *buf, size_t count,
 }
 
 #ifdef CONFIG_PROC_FS
-static int timerfd_show(struct seq_file *m, struct file *file)
+static void timerfd_show(struct seq_file *m, struct file *file)
 {
 	struct timerfd_ctx *ctx = file->private_data;
 	struct itimerspec t;
@@ -298,18 +298,19 @@ static int timerfd_show(struct seq_file *m, struct file *file)
 	t.it_interval = ktime_to_timespec(ctx->tintv);
 	spin_unlock_irq(&ctx->wqh.lock);
 
-	return seq_printf(m,
-			  "clockid: %d\n"
-			  "ticks: %llu\n"
-			  "settime flags: 0%o\n"
-			  "it_value: (%llu, %llu)\n"
-			  "it_interval: (%llu, %llu)\n",
-			  ctx->clockid, (unsigned long long)ctx->ticks,
-			  ctx->settime_flags,
-			  (unsigned long long)t.it_value.tv_sec,
-			  (unsigned long long)t.it_value.tv_nsec,
-			  (unsigned long long)t.it_interval.tv_sec,
-			  (unsigned long long)t.it_interval.tv_nsec);
+	seq_printf(m,
+		   "clockid: %d\n"
+		   "ticks: %llu\n"
+		   "settime flags: 0%o\n"
+		   "it_value: (%llu, %llu)\n"
+		   "it_interval: (%llu, %llu)\n",
+		   ctx->clockid,
+		   (unsigned long long)ctx->ticks,
+		   ctx->settime_flags,
+		   (unsigned long long)t.it_value.tv_sec,
+		   (unsigned long long)t.it_value.tv_nsec,
+		   (unsigned long long)t.it_interval.tv_sec,
+		   (unsigned long long)t.it_interval.tv_nsec);
 }
 #else
 #define timerfd_show NULL
diff --git a/include/linux/fs.h b/include/linux/fs.h
index a4ce5bae..e2f67f0 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1494,7 +1494,7 @@ struct file_operations {
 	int (*setlease)(struct file *, long, struct file_lock **, void **);
 	long (*fallocate)(struct file *file, int mode, loff_t offset,
 			  loff_t len);
-	int (*show_fdinfo)(struct seq_file *m, struct file *f);
+	void (*show_fdinfo)(struct seq_file *m, struct file *f);
 };
 
 struct inode_operations {
-- 
1.8.1.2.459.gbcd45b4.dirty

Re: [PATCH 5/7] fs: Convert show_fdinfo functions to void

[Steven Rostedt]

On Mon, 29 Sep 2014 16:08:25 -0700
Joe Perches <joe@perches.com> wrote:

> seq_printf functions shouldn't really check the return value.
> Checking seq_is_full occasionally is used instead.
> 
> Update vfs documentation.
> 
> Signed-off-by: Joe Perches <joe@perches.com>


>  
> -#ifdef CONFIG_PROC_FS
> -static int eventfd_show_fdinfo(struct seq_file *m, struct file *f)
> +static void eventfd_show_fdinfo(struct seq_file *m, struct file *f)
>  {
> +#ifdef CONFIG_PROC_FS
>  	struct eventfd_ctx *ctx = f->private_data;
> -	int ret;
>  
>  	spin_lock_irq(&ctx->wqh.lock);
> -	ret = seq_printf(m, "eventfd-count: %16llx\n",
> -			 (unsigned long long)ctx->count);
> +	seq_printf(m, "eventfd-count: %16llx\n",
> +		   (unsigned long long)ctx->count);
>  	spin_unlock_irq(&ctx->wqh.lock);
> -
> -	return ret;
> -}
>  #endif
> +}
>  
>  static const struct file_operations eventfd_fops = {
> -#ifdef CONFIG_PROC_FS
>  	.show_fdinfo	= eventfd_show_fdinfo,
> -#endif

I wouldn't change logic on this. There's no reason to call this
function if it isn't doing anything.

I'll change this to just do the update and not change logic like this.

-- Steve

>  	.release	= eventfd_release,
>  	.poll		= eventfd_poll,
>  	.read		= eventfd_read,



> diff --git a/fs/proc/fd.c b/fs/proc/fd.c
> index e11d7c5..4c3c253 100644
> --- a/fs/proc/fd.c
> +++ b/fs/proc/fd.c
> @@ -53,7 +53,7 @@ static int seq_show(struct seq_file *m, void *v)
>  			   (long long)file->f_pos, f_flags,
>  			   real_mount(file->f_path.mnt)->mnt_id);
>  		if (file->f_op->show_fdinfo)
> -			ret = file->f_op->show_fdinfo(m, file);
> +			file->f_op->show_fdinfo(m, file);
>  		fput(file);
>  	}
>  

Re: [PATCH 5/7] fs: Convert show_fdinfo functions to void

[Joe Perches]

On Tue, 2014-10-28 at 10:11 -0400, Steven Rostedt wrote:
> On Mon, 29 Sep 2014 16:08:25 -0700
> Joe Perches <joe@perches.com> wrote:
> 
> > seq_printf functions shouldn't really check the return value.
> > Checking seq_is_full occasionally is used instead.
> > 
> > Update vfs documentation.
> > 
> > Signed-off-by: Joe Perches <joe@perches.com>
> 
> 
> >  
> > -#ifdef CONFIG_PROC_FS
> > -static int eventfd_show_fdinfo(struct seq_file *m, struct file *f)
> > +static void eventfd_show_fdinfo(struct seq_file *m, struct file *f)
> >  {
> > +#ifdef CONFIG_PROC_FS
> >  	struct eventfd_ctx *ctx = f->private_data;
> > -	int ret;
> >  
> >  	spin_lock_irq(&ctx->wqh.lock);
> > -	ret = seq_printf(m, "eventfd-count: %16llx\n",
> > -			 (unsigned long long)ctx->count);
> > +	seq_printf(m, "eventfd-count: %16llx\n",
> > +		   (unsigned long long)ctx->count);
> >  	spin_unlock_irq(&ctx->wqh.lock);
> > -
> > -	return ret;
> > -}
> >  #endif
> > +}
> >  
> >  static const struct file_operations eventfd_fops = {
> > -#ifdef CONFIG_PROC_FS
> >  	.show_fdinfo	= eventfd_show_fdinfo,
> > -#endif
> 
> I wouldn't change logic on this. There's no reason to call this
> function if it isn't doing anything.
> 
> I'll change this to just do the update and not change logic like this.

Fewer #ifdefs

Re: [PATCH 5/7] fs: Convert show_fdinfo functions to void

[Steven Rostedt]

On Tue, 28 Oct 2014 07:31:32 -0700
Joe Perches <joe@perches.com> wrote:

 
> > I wouldn't change logic on this. There's no reason to call this
> > function if it isn't doing anything.
> > 
> > I'll change this to just do the update and not change logic like this.
> 
> Fewer #ifdefs
> 

And there's other ways to fix it (like using an #else), but that is
off-topic to the current change set. In other words, that change should
be separate, as I don't want discussions on what's the best use of
#ifdefs distracting from getting in the "void seq_printf()" changes.

-- Steve

Re: [PATCH 0/7] seq_printf cleanups

[Steven Rostedt]

Joe,

If you haven't already done so, can you update checkpatch.pl to
complain if someone checks the return value of seq_printf(),
seq_puts(), or seq_putc().

It should state that those functions will soon be returning void.

Thanks!

-- Steve

Re: [PATCH 0/7] seq_printf cleanups

[Joe Perches]

On Tue, 2014-10-28 at 11:32 -0400, Steven Rostedt wrote:
> If you haven't already done so, can you update checkpatch.pl to
> complain if someone checks the return value of seq_printf(),
> seq_puts(), or seq_putc().

I'm not sure that matters much as a rule because I
hope soon the compiler will bleat when that happens.

There are several more too:

	seq_vprintf
	seq_escape
	seq_write
	seq_bitmap
	seq_cpumask/seq_nodemask (and _list variants),
	seq_put_decimal_xxx